gut microbiome affects the response to anti-pd-1 ...encyclopedia of genes and genomes (kegg)...

SHORT REPORT Open Access

Gut microbiome affects the response toanti-PD-1 immunotherapy in patients withhepatocellular carcinomaYi Zheng1†, Tingting Wang2†, Xiaoxuan Tu1, Yun Huang2, Hangyu Zhang1, Di Tan2, Weiqin Jiang1, Shunfeng Cai2,Peng Zhao1, Ruixue Song2, Peilu Li2, Nan Qin2,3* and Weijia Fang1,4,5,6*

Abstract

Background: Checkpoint-blockade immunotherapy targeting programmed cell death protein 1 (PD-1) has recentlyshown promising efficacy in hepatocellular carcinoma (HCC). However, the factors affecting and predicting theresponse to anti-PD-1 immunotherapy in HCC are still unclear. Herein, we report the dynamic variationcharacteristics and specificities of the gut microbiome during anti-PD-1 immunotherapy in HCC using metagenomicsequencing.

Results: Fecal samples from patients responding to immunotherapy showed higher taxa richness and more genecounts than those of non-responders. For dynamic analysis during anti-PD-1 immunotherapy, the dissimilarity ofbeta diversity became prominent across patients as early as Week 6. In non-responders, Proteobacteria increasedfrom Week 3, and became predominant at Week 12. Twenty responder-enriched species, including Akkermansiamuciniphila and Ruminococcaceae spp., were further identified. The related functional genes and metabolic pathwayanalysis, such as carbohydrate metabolism and methanogenesis, verified the potential bioactivities of responder-enriched species.

Conclusions: Gut microbiome may have a critical impact on the responses of HCC patients treated with anti-PD-1immunotherapy. The dynamic variation characteristics of the gut microbiome may provide early predictions of theoutcomes of immunotherapy in HCC, which is critical for disease-monitoring and treatment decision-making.

Keywords: Gut microbiome, Hepatocellular carcinoma, Anti-PD-1 immunotherapy

IntroductionWith a high malignancy and a poor prognosis, hepatocellu-lar carcinoma (HCC) is ranked as the fourth-leading causeof cancer-related fatalities worldwide, bringing a heavy bur-den to public health [1]. The oral multi-kinase inhibitor, so-rafenib, is the standard systemic therapeutic option foradvanced-stage HCC and has an objective response rate(ORR) of less than 5% [2]. Recently, checkpoint blockadeimmunotherapy targeting programmed cell death protein 1(PD-1) has shown promising efficacy for treating HCC. Intwo multicenter, phase 2 studies [3, 4] investigating the

efficacy of anti-PD-1 immunotherapy in sorafenib-refractory HCC, the ORR was nearly 20%, which quadru-pled that of sorafenib. However, the factors affecting andpredicting the response to anti-PD-1 immunotherapy inHCC are still unclear.The role of the gut microbiome in modulating tumor

responses to immunotherapy in melanoma [5–7], non-small cell lung cancer, renal cell carcinoma, and urothe-lial carcinoma [6, 8] has received increased attentionacross a series of studies in recent years. However, dataconcerning the impact of the gut microbiome on HCCimmunotherapy have not been reported. Additionally,previous research to date has tended to focus more onthe baseline status rather than a dynamic variation ofthe gut microbiome during immunotherapy. The aim ofthe present study, using fecal metagenomics as a

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

* Correspondence: [email protected]; [email protected]†Yi Zheng and Tingting Wang contributed equally to this work.2Realbio Genomics Institute, 138 Xinjunhuan Road, Shanghai, China1Cancer Biotherapy Center, First Affiliated Hospital, School of Medicine,Zhejiang University, 79 Qingchun Road, Hangzhou, ChinaFull list of author information is available at the end of the article

Zheng et al. Journal for ImmunoTherapy of Cancer (2019) 7:193 https://doi.org/10.1186/s40425-019-0650-9

on July 1, 2021 by guest. Protected by copyright.

http://jitc.bmj.com

/J Im

munother C

ancer: first published as 10.1186/s40425-019-0650-9 on 23 July 2019. Dow

nloaded from

http://crossmark.crossref.org/dialog/?doi=10.1186/s40425-019-0650-9&domain=pdfhttp://orcid.org/0000-0001-9849-347Xhttp://creativecommons.org/licenses/by/4.0/http://creativecommons.org/publicdomain/zero/1.0/mailto:[email protected]:[email protected]://jitc.bmj.com/

snapshot, was to provide more specific understandingson how the gut microbiome influences the responses ofHCC patients to anti-PD-1 immunotherapy.

Results and discussionEight HCC patients with Barcelona Clinic Liver Cancer(BCLC) Stage C disease treated with anti-PD-1 anti-bodies after progression on sorafenib were enrolled inthe present study. Anti-PD-1 antibodies were adminis-tered every 3 weeks. During the treatment, no antibioticswere applied. Patients were classified as responders (R,complete or partial response, or stable disease lasting forover 6 months; n = 3) and non-responders (NR, pro-gressed disease or stable disease lasting less than 6months; n = 5) based on radiological evaluation accord-ing to Response Evaluation Criteria in Solid Tumors(RECIST 1.1). Fecal samples were collected at the base-line (Day 0), Week 1 after treatment initiation, and every3 weeks during therapy until disease progression, ac-cording to the informed consent and study protocol. Dy-namic variation of gut bacterial characteristics wasevaluated and analyzed by metagenomic sequencing.

Over the entire treatment, R showed higher taxa rich-ness and more gene counts than those of NR (Fig. 1a).As for dynamic diversity analysis, the beta diversity eval-uated by Bray-Curtis distances showed that the inter-group dissimilarity became significantly higher than theintra-group differentiation as early as Week 6 (Fig. 1band Additional file 1: Figure S1). Dynamic microbialcomposition variation was also analyzed. Before treat-ment initiation, the Gram-positive Firmicutes, Gram-negative Bacteroidetes and Gram-negative Proteobacteriadominated the fecal microbiome of both R and NR,which was in accordance with the findings in healthyadults [9], suggesting that no severe gut microbiomedysbiosis was present in the study group at baseline.Specifically, Bacteroidetes were most abundant, followedby Firmicutes and Proteobacteria. As treatment pro-ceeded, microbial composition at the phylum level in Rremained relatively stable. However, in NR, Proteobac-teria markedly increased as early as Week 3, and becamepredominant at Week 12 (Fig. 1c). Increase of Proteobac-teria in NR was mainly attributed to the prevalence ofEscherichia coli, while the most notable proteobacterialmember in R was Klebsiella pneumoniae. Composition

Fig. 1 Difference in microbial diversity and composition between R and NR. a Alpha diversity measurements by species richness (up) and genecounts (down). Red: R; Blue: NR. b Beta diversity measurements, as indicated by intra- (orange) and inter-group (green) Bray-Curtis distances. cMicrobial composition of R (left) and NR (right) at the phylum level. The ten most abundant phyla of each group are shown

Zheng et al. Journal for ImmunoTherapy of Cancer (2019) 7:193 Page 2 of 7


http://jitc.bmj.com

/J Im

munother C


nloaded from

http://jitc.bmj.com/

of Bacteroides and Firmicutes also exhibited individualpatterns (Additional file 2: Figure S2 and Additional file 3:Figure S3). These findings suggested that the dynamicvariation characteristics of gut-microbial diversity andcomposition at the early treatment period of anti-PD-1immunotherapy in HCC may have distinct implicationson drug efficacy and disease prognosis.To further identify the species possibly affecting pa-

tient responses, linear discriminant analysis (LDA)-effectsize (LEfSe) algorithm analysis was performed betweenall R and NR samples. Twenty R-enriched species andfifteen NR-enriched species were identified (Fig. 2a andAdditional file 4: Figure S4). Among the R-enriched spe-cies, four Lactobacillus species (L. oris, L. mucosae, L.gasseri, and L. vaginalis), Bifidobacterium dentium andStreptococcus thermophilus were probiotic lactic acidbacteria, which were beneficial to host metabolism andimmunity by inhibiting the growth of pathogenic

microorganisms and accompanying spoilage agents. Oraladministration of Bifidobacterium could improve tumorcontrol efficacy of programmed cell death protein 1 ligand1 (PD-L1)-specific antibody therapy [10]; Coprococcuscomes, Bacteroides cellulosilyticus, and Subdoligranulum sp.also possessed probiotic potential, as they were reported tobe related with dietary fiber digestion and short-chain fattyacid production. Notably, enrichment of one Lachnospira-ceae and two Ruminococcaceae species (Lachnospiraceaebacterium 7_1_58FAA, Ruminococcus obeum, Ruminococ-cus bromii) and Akkermansia muciniphila in R was also ob-served. Commensal A.muciniphila and Ruminococcaceaebenefited host health by preventing increases in intestinalpermeability and systemic immunosuppression. In previousstudies, a significantly higher relative abundance of Rumi-nococcaceae was identified in melanoma patients respond-ing to anti-PD-1 immunotherapy [7], and oralsupplementation with A. muciniphila could restore the

Fig. 2 Meta-analysis of the bacteria significantly enriched in R and NR. a Heatmap showing the relative abundance of R-enriched and NR-enrichedbacterial species, as identified by LEfSe. b Correlation network of the R-enriched and NR-enriched species (Spearman correlations with rho> 0.5, P< 0.01were shown). The size of the nodes is proportional to the averaged relative abundance of these species in all samples. The thicknesses of the lines denotethe strengths of the correlations. c Positive correlation network of the significant R-enriched species and KO categories



http://jitc.bmj.com

/J Im

munother C


nloaded from


efficacy of anti-PD-1 immunotherapy [8]. In the presentstudy, the SparCC algorithm was applied to gain insightsinto the mutualistic networks between R-enriched and NR-enriched species (Fig. 2b). The number of significantpositive-correlation pairs and the correlation strengths inR-enriched species were higher than those of NR-enrichedspecies. Among the R-enriched bacteria, the four Lactoba-cillus species were most significantly correlated with eachother, indicating their possible pivotal roles in the network.Our findings further indicated the biological significance ofparticular bacterial strains during anti-PD-1 immunother-apy in HCC and may provide support for the developmentof a gut-microbiome-modulation scheme inimmunotherapy.Next, functional gene families associated with R-

enriched bacteria were investigated. A total of 189 KyotoEncyclopedia of Genes and Genomes (KEGG) ortholo-gies (KOs) were identified to be enriched in R (Kruskal-Wallis rank sum test, P < 0.05). Significant positive cor-relations were identified between 123 R-enriched KOsand 18 R-enriched species (Spearman’s correlation, rho >0.5, Additional file 6: Table S1). Pathway identification ofKOs verified the potential bioactivities of R-enrichedspecies (Fig. 2c and Additional file 5: Figure S5). In de-tail, the cellobiose-transport system (ko02010) was sig-nificantly correlated with B. dentium; the pectin lyase(K01732), which may be involved in pectin metabolism,was correlated with A. muciniphila. Both cellulose andpectin have been highlighted for their prebiotic and anti-inflammatory potential as dietary fibers [11, 12]. Meth-anogenesis pathway (ko00680) was found to be corre-lated with R. obeum and the four Lactobacillus species.Methane generated in human gastrointestinal tract hasbeen reported to ameliorate oxidative stress injury andsuppress the host inflammatory response [13]. Otherpathways with potential benefits included sulfate reduc-tion (ko00920) and carbon fixation (ko00720) functionsthat were correlated with R. obeum, carotenoid biosyn-thesis (ko00906) correlated with B. cellulosilyticus and A.colihominis, and unsaturated fatty acid metabolism(ko00590) associated with C. comes (Additional file 7:Table S2). Such findings further illustrated the potentialunderlying mechanisms of gut microbiome influencingthe anti-PD-1 immunotherapy efficacy in HCC patients.It is well recognized that factors including age, genet-

ics, and diet may influence microbiome composition[14]. However, the long-term stability of the human gutmicrobiota has been demonstrated in previous studies.A study demonstrated that an individual’s microbiotacould be remarkably stable, with 60% of strainsremaining over the course of five years; this findinghighlighted that such stability and responsiveness tophysiologic change confirmed the potential of the gutmicrobiota as a diagnostic tool and therapeutic target

[15]. In the present study, patients were enrolled withstrict criteria to minimize potential dietary or geograph-ical influences during the whole course of treatment, aswell as to ensure that the dynamic changes of gut micro-biota in both R and NR were attributed to the thera-peutic intervention instead of any daily factors.Although our present microbiome-wide association

study highlighted the association between changes in thegut microbiome and host-immune responses to drugs,any causal relationship of these correlative associationsremains unknown. To further elucidate this relationship,several strategies have been developed. Firstly, ‘meta-omic’ analysis combining sequencing and multiple bio-chemical techniques can greatly advance the knowledgeof the human microbiome and its specific role in gov-erning disease states. For example, Gopalakrishnan et al.compared the tumor-associated immune infiltrates viamulti-parameter immunohistochemistry (IHC) andfound a statistically significant positive correlation be-tween CD8+ T cell infiltrate in the tumor and the Faeca-libacterium genus as well as the Ruminococcaceaefamily, suggesting a possible mechanism through whichthe gut microbiome may modulate anti-tumor immuneresponses [7]. More importantly, in vitro or in vivo ex-perimental models are required in order to allow thesystematic manipulation of variables and, thus, allow ex-perimental testing and validation of results derived frommeta-omics [16]. In our previous study, we had demon-strated that Behcet’s disease (BD), a kind of autoimmunedisorder, was associated with considerable gut micro-biome changes [17]. To determine whether the gutmicrobiome contributes to development of this disease,fecal microbiota transplantation (FMT) was performedin mice undergoing autoimmune uveitis. We demon-strated that mice colonized with whole-gut microbiomesfrom BD patients showed an exacerbation of disease ac-tivity and an excessive production of pro-inflammatorycytokines. These results may confirm the hypothesis thatspecific bacterial patterns contributed to the develop-ment of intraocular inflammatory disease. In our futurestudies, similar meta-omics strategies, as well as the useof mouse FMT models, will also be incorporated inorder to deeply explore the interaction between the gutmicrobiome and host responses to immunotherapiesamong HCC patients.Gopalakrishnan et al. also assessed the landscape of

the oral and gut microbiomes in patients with metastaticmelanoma receiving anti-PD-1 immunotherapy anddemonstrated a high abundance of Lactobacillales in theoral microbiomes compared with that of fecal micro-biomes in all subjects [7]. Galloway-Pena et al. furtherreported a high degree of intra-patient temporal instabil-ity of both stool and oral microbial diversity in patientswith acute myeloid leukemia undergoing induction



http://jitc.bmj.com

/J Im

munother C


nloaded from


chemotherapy [18]. Both studies showed the possibilityof using oral microbiome, rather than fecal microbiome,as an indicator for specific clinical outcomes. In our fu-ture longitudinal studies, both fecal and oral microbiomewill be taken into consideration for a more comprehen-sive visualization of the host-microbe interactions.In conclusion, by metagenomic sequencing of periodic

fecal samples, we illustrated that the dynamic-variationcharacteristics of the gut microbiome might be used forearly prediction of the six-month outcomes of anti-PD-1immunotherapy in HCC at 3–6 weeks after treatmentinitiation, which is critical for disease monitoring andtreatment decision-making. To our knowledge, this isthe first study to focus on the association between thegut microbiome and the response to anti-PD-1 immuno-therapy for HCC. Furthermore, while previous studieshad mostly only provided cross-sectional comparisons,the present study showed the different trajectories ofmicrobiome shifts between R and NR and revealed astronger stability of the gut microbiome in R during thewhole course of treatment. We believe that the potentialof the gut microbiota as a therapeutic target is indicatedin this study, and that the relevant bacterial species andmetabolic pathways revealed here could be developed asa modulation strategy for better treatment options forHCC patients.

MethodsPatients and medicationsA cohort of eight patients was included in this study. Allpatients were histologically confirmed HCC with BCLCStage C disease and had experienced disease progressionon first-line sorafenib treatment. Other eligibility criteriaincluded radiological measurable disease, an adequatefunction of major organs, an Eastern Cooperative Oncol-ogy Group (ECOG) performance status of 0 or 1, and aChild-Pugh Class A liver function. Exclusion criteria in-cluded fibrolamellar HCC, sarcomatoid HCC, or mixedcholangiocarcinoma and HCC, liver transplantation, or ahistory of active autoimmune disease.Patients received camrelizumab (SHR-1210, HengRui

Medicine Co., Jiangsu, China), a humanized anti-PD-1IgG4 monoclonal antibody, intravenously at a dose of 3mg/kg every three weeks until disease progression or in-tolerance (Clinicaltrials.gov ID: NCT02989922). Writteninformed consent, including collection and analysis ofmicrobiome samples, was obtained from each patientbefore initiation of the study. Patients were asked tokeep their own dietary and other habits during drugtreatment to avoid any interruption to their inherent gutmicrobiome. Additionally, in order to omit the geograph-ical influence, we collected samples only from one hospitalwith local individuals who shared common dietary intake

patterns. None of the patients had experienced diarrhea orother intestinal symptoms, or antibiotic/probiotic con-sumption during the study.Radiological evaluation was assessed every six weeks

according to the RECIST 1.1 criteria. Responders (R,n = 3) were defined by radiographic evidence ascomplete or partial response, or stable disease lasting forat least six months. Non-responders (NR, n = 5) weredefined as those with progressed disease or stable dis-ease lasting less than six months. This study compliedwith the Declaration of Helsinki and was approved bythe Ethics Committee of the First Affiliated Hospital ofZhejiang University.

Fecal sample collectionFresh feces were collected on the last pre-treatmentclinic visit (as baseline, Day 0), one week after treatmentinitiation, the day of each treatment and were stored at− 80 °C immediately prior to DNA extraction. Accordingto the sample providing policy in the informed consent,the observation lasted for 39, 21, and 18 weeks for thethree responders, and 6, 6, 9, 9, and 12 weeks for the fivenon-responders.

DNA extraction and metagenomic sequencingDynamic variation of gut bacterial composition of R andNR during the anti-PD-1 immunotherapy was evaluatedusing fecal metagenomic sequencing. Briefly, bacterialgenomic DNA was extracted using QIAamp DNA StoolMini Kit (Qiagen, Hilden, Germany). After checking theintegrity and concentration of DNA, individual librarieswere constructed using MGIEasy DNA Library Prep Kit(BGI, Shenzhen, China), loaded into the BGISEQ-500 RSplatform (BGI, Shenzhen, China), and were sequencedusing 2 × 100 bp paired-end read protocol. The processof quality filtering, trimming, and demultiplexing wascarried out as described previously [19]. A total of 49datasets (28 from R and 21 from NR) were generated.Overall, 78.12% of the raw reads were regarded as high-quality reads, with an average length of 72 bp and anaverage Q35 score of 100% (Additional file 8: Table S3).

Taxonomic and gene profilingAll high-quality reads were then aligned to Homo sapiens(human) genome assembly hg38 [20] using SOAPalign 2.21with default parameters to remove human reads (https://anaconda.org/bioconda/soapaligner). The retained cleanreads were then aligned to ~ 1M clade-specific markergenes from approximately 17,000 reference genomes for es-timation of relative phylotype abundance using MetaPhlAn(version 2.5.0) [21].For gene annotation, the clean reads were aligned to

the integrated gene catalog (IGC) [22] by using SOAPa-lign 2.21 with default parameters; only the reads with



http://jitc.bmj.com

/J Im

munother C


nloaded from

http://clinicaltrials.govhttps://anaconda.org/bioconda/soapalignerhttps://anaconda.org/bioconda/soapalignerhttp://jitc.bmj.com/

both ends mapped to the same gene were used in thefollowing analysis. An average IGC mapping rate of77.77% and an average unique mapping rate of 63.27%were achieved. The gene relative abundance profile wasgenerated following the procedure described before [17].Functional annotations were carried out by BLASTPsearch against the KEGG database (e value ≤1e − 5 andhigh-scoring segment pair scoring > 60) [23]. The KOabundance was estimated by accumulating the relativeabundance of all genes belonging to this feature.

Statistical analyses and correlation networkNon-parametric Wilcoxon rank-sum test was employedto analyze the statistical significance of diversity indices,taxa and KOs between R and NR. Bray-Curtis metricswere applied to calculate pairwise dissimilarities betweensamples and were used in beta diversity evaluation andprincipal coordinate analysis (PCoA) [24]. The LEfSe al-gorithm was applied to identify the phylotypes signifi-cantly different in relative abundance between all R andNR samples; phylotypes with an LDA score cut-off of2.0 and P < 0.05 in built-in rank sum test were regardedas statistically significant [25].SparCC algorithm was used to calculate correlations

between R-enriched and NR-enriched species. Boot-strapping of 100 repetitions was used to compute the Pvalue for each correlation, as described previously [17].Only significant correlations with P < 0.05 and rho > 0.5were presented in the network. Spearman correlationwas applied to estimate the association strengths for de-termining the relationship between bacteria and KO cat-egories. Only significant correlations with P < 0.01 andrho > 0.5 were presented in the network. The species-species network and the species-KO network were bothvisualized with Cytoscape3.0.2.

Additional files

Additional file 1: Figure S1. Principal componeLnt analysis (PCoA)based on Bray-Curtis distances, followed by Adonis test at Day 0 (a), Week1 (b), Week 3 (c), and Week 6 (d). (PNG 1455 kb)

Additional file 2: Figure S2. Microbial composition of the R (left) andNR (right) at the genus (a) and species (b) levels. The ten most abundantgenera or species of each group are shown. (PNG 4920 kb)

Additional file 3: Figure S3. Sankey analysis of all R and NR during thetreatment. (PNG 9946 kb)

Additional file 4: Figure S4. Differentially abundant genera (a) andspecies (b) between R and NR, identified by LEfSe. (PNG 2863 kb)

Additional file 5: Figure S5. Positive correlation network of significantR-enriched species and KOs. Correlations with rho > 0.7 are shown in red,while those with rho > 0.5 are shown in gray. (PNG 3679 kb)

Additional file 6: Table S1. List of significant positive correlationsbetween species and KOs. (XLSX 27 kb)

Additional file 7: Table S2. Pathway assignment of the mostsignificantly correlated KOs in R. (XLSX 14 kb)

Additional file 8: Table S3. Quality control of the sequencing data.(XLSX 19 kb)

AbbreviationsBCLC: Barcelona Clinic Liver Cancer; ECOG: Eastern Cooperative OncologyGroup; HCC: hepatocellular carcinoma; IGC: integrated gene catalog;KEGG: Kyoto Encyclopedia of Genes and Genomes; KOs: Kyoto Encyclopediaof Genes and Genomes Orthologies; LDA: Linear discriminant analysis;LEfSe: Linear discriminant analysis - effect size; NR: Non-responders;ORR: Objective response rate; PCoA: Principal coordinate analysis; PD-1: Programmed cell death protein 1; PD-L1: Programmed cell death protein 1ligand 1; R: Responders; RECIST 1.1: Response Evaluation Criteria in SolidTumors 1.1

AcknowledgementsWe would like to thank the patients for their participation. We are grateful toall the participants who have made this research possible.

Authors’ contributionsYZ, TW, SC, NQ and WF designed the study. YZ, TW, XT, HZ, WJ, PZ and PLcollected samples and completed the laboratory experiments. TW, YH, DTand RS performed the statistical analyses. NQ and WF led the study. YZ, TW,XT, SC and DT drafted the manuscript. All authors reviewed the manuscript,provided feedback, and approved the manuscript in its final form.

FundingDr. NQ was supported by the National Natural Science Foundation of China(31670118) and the Natural Science Foundation of Zhejiang Province(LR15H030002). Dr. WF was supported by the Key Subject of Clinical MedicalEngineering of Zhejiang Province (No. G3221), S&T Major Project of China(2017ZX10203205), and the Major Scientific Project of Zhejiang Province(2017C03028).

Availability of data and materialsAll sequencing datasets were uploaded to the NCBI database with SRAaccession: PRJNA505228.

Ethics approval and consent to participateThis study was approved by the Ethics Committee of the First AffiliatedHospital of Zhejiang University. All subjects provided informed consent to beincluded in the study.

Consent for publicationNot applicable.

Competing interestsThe authors declare that they have no competing interests.

Author details1Cancer Biotherapy Center, First Affiliated Hospital, School of Medicine,Zhejiang University, 79 Qingchun Road, Hangzhou, China. 2Realbio GenomicsInstitute, 138 Xinjunhuan Road, Shanghai, China. 3Shanghai Tenth People’sHospital Affiliated to Tongji University, No. 301 Yanchangzhong Road,Shanghai, China. 4Key Laboratory for Drug Evaluation and Clinical Researchof Zhejiang Province, First Affiliated Hospital, School of Medicine, ZhejiangUniversity, 79 Qingchun Road, Hangzhou, China. 5Key Laboratory of PrecisionDiagnosis and Treatment for Hepatobiliary and Pancreatic Tumor of ZhejiangProvince, First Affiliated Hospital, School of Medicine, Zhejiang University, 79Qingchun Road, Hangzhou, China. 6Zhejiang Provincial Key Laboratory ofPancreatic Disease, First Affiliated Hospital, School of Medicine, ZhejiangUniversity, 79 Qingchun Road, Hangzhou, China.

Received: 26 February 2019 Accepted: 21 June 2019

References1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin. 2018;

68(1):7–30.2. Cheng AL, Kang YK, Chen Z, Tsao CJ, Qin S, Kim JS, et al. Efficacy and safety

of sorafenib in patients in the Asia-Pacific region with advanced



http://jitc.bmj.com

/J Im

munother C


nloaded from

https://doi.org/10.1186/s40425-019-0650-9https://doi.org/10.1186/s40425-019-0650-9https://doi.org/10.1186/s40425-019-0650-9https://doi.org/10.1186/s40425-019-0650-9https://doi.org/10.1186/s40425-019-0650-9https://doi.org/10.1186/s40425-019-0650-9https://doi.org/10.1186/s40425-019-0650-9https://doi.org/10.1186/s40425-019-0650-9http://jitc.bmj.com/

hepatocellular carcinoma: a phase III randomised, double-blind, placebo-controlled trial. The Lancet Oncology. 2009;10(1):25–34.

3. El-Khoueiry AB, Sangro B, Yau T, Crocenzi TS, Kudo M, Hsu C, et al.Nivolumab in patients with advanced hepatocellular carcinoma(CheckMate 040): an open-label, non-comparative, phase 1/2 doseescalation and expansion trial. Lancet (London, England). 2017;389(10088):2492–502.

4. Zhu AX, Finn RS, Edeline J, Cattan S, Ogasawara S, Palmer D, et al.Pembrolizumab in patients with advanced hepatocellular carcinomapreviously treated with sorafenib (KEYNOTE-224): a non-randomised,open-label phase 2 trial. The Lancet Oncology. 2018;19(7):940–52.

5. Matson V, Fessler J, Bao R, Chongsuwat T, Zha Y, Alegre ML, et al.The commensal microbiome is associated with anti-PD-1 efficacy inmetastatic melanoma patients. Science. 2018;359(6371):104.

6. Vetizou M, Pitt JM, Daillere R, Lepage P, Waldschmitt N, Flament C, et al. Anticancerimmunotherapy by CTLA-4 blockade relies on the gut microbiota. Science. 2015;350(6264):1079–84.

7. Gopalakrishnan V, Spencer CN, Nezi L, Reuben A, Andrews MC, Karpinets TV, et al.Gut microbiome modulates response to anti-PD-1 immunotherapy in melanomapatients. Science. 2018;359(6371):97–103.

8. Routy B, Le Chatelier E, Derosa L, Duong CPM, Alou MT, Daillere R, et al. Gutmicrobiome influences efficacy of PD-1-based immunotherapy against epithelialtumors. Science. 2018;359(6371):91–7.

9. Human Microbiome Project C. Structure, function and diversity of the healthyhuman microbiome. Nature. 2012;486(7402):207–14.

10. Sivan A, Corrales L, Hubert N, Williams JB, Aquino-Michaels K, Earley ZM, et al.Commensal Bifidobacterium promotes antitumor immunity and facilitates anti-PD-L1 efficacy. Science. 2015;350(6264):1084–9.

11. Shoaie S, Ghaffari P, Kovatcheva-Datchary P, Mardinoglu A, Sen P,Pujos-Guillot E, et al. Quantifying diet-induced metabolic changes ofthe Human gut Microbiome. Cell Metab. 2015;22(2):320–31.

12. Chung WSF, Meijerink M, Zeuner B, Holck J, Louis P, Meyer AS, et al.Prebiotic potential of pectin and pectic oligosaccharides to promoteanti-inflammatory commensal bacteria in the human colon. FEMSMicrobiol Ecol. 2017;1, 93(11).

13. Zhang X, Li N, Shao H, Meng Y, Wang L, Wu Q, et al. Methane limitLPS-induced NF-kappaB/MAPKs signal in macrophages and suppressimmune response in mice by enhancing PI3K/AKT/GSK-3beta-mediated IL-10 expression. Sci Rep. 2016;6:29359.

14. Wu GD, Chen J, Hoffmann C, Bittinger K, Chen YY, Keilbaugh SA, etal. Linking long-term dietary patterns with gut microbial enterotypes.Science. 2011;334(6052):105–8.

15. Faith JJ, Guruge JL, Charbonneau M, Subramanian S, Seedorf H,Goodman AL, et al. The long-term stability of the human gutmicrobiota. Science. 2013;341(6141):1237439.

16. Fritz JV, Desai MS, Shah P, Schneider JG, Wilmes P. From meta-omicsto causality: experimental models for human microbiome research.Microbiome. 2013;1(1):14.

17. Ye Z, Zhang N, Wu C, Zhang X, Wang Q, Huang X, et al. Ametagenomic study of the gut microbiome in Behcet's disease.Microbiome. 2018;6(1):135.

18. Galloway-Pena JR, Smith DP, Sahasrabhojane P, Wadsworth WD,Fellman BM, Ajami NJ, et al. Characterization of oral and gutmicrobiome temporal variability in hospitalized cancer patients.Genome medicine. 2017;9(1):21.

19. Fang C, Zhong H, Lin Y, Chen B, Han M, Ren H, et al. Assessment of thecPAS-based BGISEQ-500 platform for metagenomic sequencing.GigaScience. 2018;7(3):1–8.

20. Rosenbloom KR, Armstrong J, Barber GP, Casper J, Clawson H,Diekhans M, et al. The UCSC genome browser database: 2015 update.Nucleic Acids Res. 2015;43(Database issue):D670–81.

21. Truong DT, Franzosa EA, Tickle TL, Scholz M, Weingart G, Pasolli E, etal. MetaPhlAn2 for enhanced metagenomic taxonomic profiling. NatMethods. 2015;12(10):902–3.

22. Li J, Jia H, Cai X, Zhong H, Feng Q, Sunagawa S, et al. An integratedcatalog of reference genes in the human gut microbiome. NatBiotechnol. 2014;32(8):834–41.

23. Kanehisa M, Goto S, Kawashima S, Okuno Y, Hattori M. The KEGGresource for deciphering the genome. Nucleic Acids Res. 2004;32(Database issue):D277–80.

24. Parks DH, Beiko RG. Measures of phylogenetic differentiation providerobust and complementary insights into microbial communities. TheISME Journal. 2013;7:173–83.

25. Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, etal. Metagenomic biomarker discovery and explanation. Genome Biol.2011;12(6):R60.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.



http://jitc.bmj.com

/J Im

munother C


nloaded from


AbstractBackgroundResultsConclusions

IntroductionResults and discussionMethodsPatients and medicationsFecal sample collectionDNA extraction and metagenomic sequencingTaxonomic and gene profilingStatistical analyses and correlation network

Additional filesAbbreviationsAcknowledgementsAuthors’ contributionsFundingAvailability of data and materialsEthics approval and consent to participateConsent for publicationCompeting interestsAuthor detailsReferencesPublisher’s Note

gut microbiome affects the response to anti-pd-1 ...encyclopedia of genes and genomes (kegg)...

Documents